A Road Network Equilibrium Control Method and System Based on Big Data

Abstract

This invention discloses a road network balancing control method and system based on big data. The method includes: acquiring and parsing the road network spatial topology, obtaining the layout of detection equipment on the road network, as well as equipment types and detectable parameters; calculating the status indicators of road network nodes such as intersections and road segments; spatially segmenting the road network into relatively independent road subnetworks based on the correlation between nodes in the road network using a clustering method; classifying the operating status of the road subnetworks according to the status indicators and calculating the carrying capacity of each road subnetwork under congested conditions; identifying key nodes inside and outside the controlled subnetworks and implementing a balancing control strategy. This invention is beneficial for balancing vehicle distribution in the road network space and improving the traffic efficiency of the road network.

Description

Technical Field

[0001] This invention belongs to the field of traffic control and big data analysis technology, and in particular, it relates to a road network equilibrium control method, system and equipment based on big data. Background Technology

[0002] With my country's economic development, the number of motor vehicles has continued to increase, exacerbating road network congestion and slowing down vehicle travel. Given limited urban space resources and high costs, simply relying on road expansion to solve the imbalance between urban traffic supply and demand is unrealistic. Therefore, solutions to this imbalance must focus on improving traffic management efficiency and optimizing traffic planning.

[0003] Traditionally, traffic management primarily focuses on the temporal changes in traffic conditions within localized areas, rarely considering large-scale road networks or the connections between roads. By passively adapting to changes in traffic demand through adjustments to traffic timing, it can alleviate supply-demand imbalances to some extent. However, this approach becomes ineffective when traffic demand exceeds the capacity of intersections, road segments, or areas due to continuous vehicle accumulation. Generally, a road network operating at full capacity across all time and space is impossible. If real-time traffic information on the road network can be obtained, and accurate traffic operation status can be grasped, scientific and reasonable traffic control measures can be implemented to guide vehicles and enable them to utilize road resources efficiently. Therefore, solutions to supply-demand imbalances should be sought from a more macroscopic perspective. Looking at the urban road network as a whole, its spatial characteristics and topology determine, to a certain extent, the distribution of traffic flow across the city. The distribution of traffic demand on the road network is uneven, and the degree of this unevenness can be accurately analyzed. Therefore, balancing the supply-demand imbalance in localized areas over a larger area can alleviate traffic problems to some extent.

[0004] Large and medium-sized cities have initially formed a big data environment based on urban vehicle travel, combining dynamic and static multi-source data. Available big data includes fixed detectors such as loop detectors, video checkpoint detectors, RFID, and microwave detectors, which can collect real-time traffic flow parameters such as flow rate, speed, occupancy, headway, and headway distance. Mobile detection equipment such as mobile phone signaling, vehicle-to-everything (V2X) data, and mobile internet data can acquire real-time traffic big data such as vehicle trajectories and driving status. This traffic data presents a more comprehensive picture of urban traffic information. The acquisition of multi-source data provides data support for traffic condition analysis and research, making it possible to explore the mechanisms of traffic congestion, monitor traffic operation in real time, and dynamically predict future traffic flow. Spatially analyzing the correlation between regions, introducing spatial structure analysis and topological feature identification methods for urban road networks, and relying on urban big data analysis technology, we can comprehensively understand the relationship between the capacity of local urban road network areas and traffic conditions, analyze the evolution of traffic conditions over time, and utilize spatial proximity and topological equivalence to identify sets of areas that influence regional traffic conditions, thus more effectively distributing traffic flow among adjacent areas. Summary of the Invention

[0005] The purpose of this invention is to address the shortcomings of existing technologies by providing a road network equilibrium control method, system, and device based on big data.

[0006] The technical solution to achieve the purpose of this invention is as follows: On the one hand, a road network equilibrium control method based on big data is provided: acquiring and parsing the spatial topology of the road network, acquiring the attribute parameters of intersections and road segments, and acquiring the layout of detection equipment on the road network, as well as the equipment type and detectable parameters;

[0007] Based on the spatial topology of the road network, calculate the status indicators of each road network node within the road network;

[0008] Based on the correlation between nodes in the road network, a clustering method is used to spatially segment the road network into different road subnetworks.

[0009] The operating status of the road subnet is classified according to the status indicators, the controlled subnet is identified, and the carrying capacity of the controlled subnet is calculated.

[0010] Based on the above segmentation results and combined with the spatial characteristics of the road network, key nodes inside and outside the controlled subnetwork are identified and control strategies are implemented.

[0011] Another aspect of this disclosure provides a method for calculating the carrying capacity of a road subnetwork. This method involves selecting appropriate state indicators for intersections and road segments within the road network; employing a multinomial fitting method to form a traffic state-on-the-road vehicle relationship function; classifying the function curve by feature, dividing the state into three intervals: smooth flow, slow flow, and congestion; using a clustering method to associate and merge intersections and road segments within the road network, segmenting them into road subnetworks; and calculating the state boundary values ​​of the road subnetworks to obtain the carrying capacity of the road subnetworks under congestion conditions.

[0012] Another aspect of this disclosure provides a method for generating a traffic state-on-the-road vehicle relationship function. This method involves selecting and analyzing traffic flow data from one week of a road network, establishing a data processing workflow, preprocessing the data to form a dataset of traffic state indicators and on-the-road vehicle numbers, and then using a polynomial fitting method to fit the traffic state-on-the-road vehicle relationship function.

[0013] In another aspect of this disclosure, a traffic operation state segmentation method is provided. Based on historical traffic data of a road network over a week, the traffic state-on-the-way vehicle relationship function is input to obtain a data sample set. The curvature of the data samples is calculated by sliding to form a curvature data sample set. A temporal clustering method is used to classify the curvature data samples to form three state intervals: smooth, slow, and congested. Thus, the range of network node carrying capacity intervals can be deduced.

[0014] Another aspect of this disclosure provides a method for calculating the correlation degree between road network nodes. A two-layer network model is constructed to represent the road network, with the upper layer representing the traffic flow distribution model and the lower layer representing the spatial structure model. A correlation degree calculation model is constructed to calculate the correlation degree matrix between road network nodes.

[0015] Another aspect of this disclosure provides a road subnet segmentation method based on graph clustering. A road network graph is constructed based on the set of congested nodes, a road subnet segmentation model is built, and a subnet set is calculated. The final road subnet set is formed by using overlap determination rules.

[0016] Another aspect of this disclosure provides a road subnet equilibrium control method. This method identifies controlled subnet subsets from a road subnet set based on historical data, calculates and sorts the external entry nodes associated with the controlled subnets, and then hierarchically processes these external entry nodes according to their association degree, forming restricted areas, buffer zones, and diversion zones. Furthermore, it acquires vehicle trajectory data in the road network in real time and calculates subnet G. iThe system collects on-the-go vehicle data; reduces green light time in restricted areas and lowers vehicle speeds in buffer zones; and provides guidance information and alternative routes in diversion areas to encourage vehicles to change their routes. Based on correlation relationships, nodes such as intersections and road segments within the subnetwork are categorized into key and non-key nodes. Real-time traffic status indicators for key nodes are calculated, aiming for balanced vehicle carrying capacity at each node. Intelligent traffic management methods are used to rapidly divert traffic flow associated with key nodes, keeping the status indicators of each node within the optimized target range.

[0017] Another aspect of this disclosure provides a road network equilibrium control system based on big data, including a data receiving module, a data processing module, a data service module, and a control optimization module. The data receiving module receives real-time traffic flow data collected by road network front-end detection equipment; the data processing module preprocesses the real-time traffic data to ensure data quality and transmits the data to the data service module; it requests historical data, combines it with spatial road network data, analyzes and processes the data, calculates status indicators, and identifies traffic conditions; the data service module serves as the system's data bus, handling data input and output and enabling data interaction with other modules; the control optimization module obtains real-time indicator data and control subnet data from the data module in real time, generates optimized control commands in real time, and sends them to the front-end execution equipment.

[0018] In another aspect, this disclosure provides an electronic device, including a memory for storing computer instructions and a processor connected to the memory for executing the computer instructions in the memory.

[0019] Compared with existing technologies, the significant advantages of this invention are: based on historical data, it identifies the functional relationship between traffic conditions and vehicles on the road network, and identifies inflection points of state changes; thereby segmenting the road network and formulating effective equilibrium control strategies based on real-time state change trends. The equilibrium control method and system disclosed herein are beneficial for balancing vehicle distribution across the road network and improving traffic efficiency.

[0020] The present invention will now be described in further detail with reference to the accompanying drawings. Attached Figure Description

[0021] Figure 1 This is a flowchart of a road network equilibrium control based on big data in one embodiment.

[0022] Figure 2 This is a flowchart of the road network model construction process in one embodiment.

[0023] Figure 3 This is a schematic diagram of the road subnet segmentation result in one embodiment.

[0024] Figure 4 This is a flowchart of road subnet segmentation in one embodiment.

[0025] Figure 5 A flowchart for generating the traffic state-on-the-way vehicle relationship function in one embodiment.

[0026] Figure 6 This is a flowchart of the state interval calculation for a relational function in one embodiment. Detailed Implementation

[0027] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0028] In one embodiment, combined Figure 1 This paper presents a road network equilibrium control method based on big data, which includes the following steps:

[0029] Step 101: Obtain and parse the road network spatial topology, obtain the attribute parameters of intersections and road segments, including: intersection shape, number of entrances and exits, lane type and number, etc.; and obtain the layout of detection equipment on the road network, as well as equipment type, detectable parameters, equipment location, etc.

[0030] Step 102: Based on the detectable traffic flow data and the spatial topology of the road network, calculate the status indicators of each intersection, road segment, and other road network nodes within the road network.

[0031] In this embodiment of the disclosure, the intersection selects single or combined indicators such as saturation, average vehicle delay, and queue length as status indicators.

[0032] In this embodiment of the disclosure, the road segment selects a single or combined indicator such as travel time, travel speed, number of stops, and saturation as the status indicator.

[0033] Step 103: Based on the correlation between nodes in the road network, a clustering method is used to spatially segment the road network into different road subnetworks. The segmentation results are as follows: Figure 3 As shown.

[0034] In this embodiment of the disclosure, a two-layer network model is designed to represent the road network. The upper layer network represents the traffic flow distribution model, and the lower layer represents the spatial structure model. Points represent intersections, and lines between points represent road segments.

[0035] In this embodiment of the disclosure, the lower-level road network topology relationship N A= (R, L, α). Where R is the set of intersection points; L represents the set of road segments connecting upstream and downstream intersections; α is the set of static correlation degrees, representing the spatial correlation between intersections within the road network. Factors affecting the correlation include: road segment length and width, traffic capacity, accessibility, etc.

[0036] In this embodiment of the disclosure, the upper-layer network is a traffic flow assignment model, and the travel network N B = (O, D, β), where O is the starting point; D is the destination point; and β is the dynamic correlation degree, which is influenced by factors such as traffic flow and travel time.

[0037] Combination Figure 2 This step specifically includes the following steps:

[0038] Step 201: According to the different control periods throughout the day, within each period, use Δt l , Δt l Slices were created at intervals of ≥15 minutes, and the slices with the most congested nodes in each time period were searched out and used as the analysis time periods.

[0039] Step 202: Obtain the set of congested nodes N for the analysis period S =(P S L S ,w), where P S For the congested intersection point set, L S Let w be the set of road segments between intersections, and w be the degree of association between nodes. Graph clustering analysis is performed on the nodes to form independent road subnetworks in the road network space.

[0040] Based on the two-layer network model, the correlation calculation model is obtained:

[0041] w i,j =α i,j *β i,j

[0042] Among them, w i,j The correlation between nodes is indicated by a value greater than 0.5, which is considered a high correlation.

[0043] Where: α i,j Static correlation degree, α i,j =1 / d i,j d i,j The average connectivity distance between nodes i and j in the routing network; β i,j Dynamic correlation, β i,j =Q ij / Q i +Q j q ij This refers to the bidirectional traffic volume between two road network nodes; Q i Q jLet be the nominal traffic volume between nodes i and j.

[0044] Of which: nominal traffic volume n refers to the number of key nodes, such as intersections and road segments; q i The weight of the selected key point; q i This corresponds to the traffic flow.

[0045] In this embodiment of the disclosure, based on the set of crowded nodes N S =(P S L S Construct a road network graph G = (P, L), where P represents all congested nodes (P1, P2, ..., Pn) in the road network. n For any two nodes in P, they are connected by an edge. The road network is bidirectional. i,j ≠w j,i The correlation matrices W' and W” are formed:

[0046]

[0047] Combination Figure 4 The method of using clustering to spatially segment the road network into different road subnetworks includes the following steps:

[0048] Step 301: Construct a road subnet segmentation model to divide the road network into unconnected road subnets:

[0049]

[0050] In the formula, the road network graph G is divided into k unconnected road subnets, and the set of road subnets is G = (G1, G2, ..., G...). k ), G i G j For the i-th and j-th road subnets, satisfying And G1∪G2∪…∪G k =G; W is W' or W”. For complement;

[0051] Step 302: Select the correlation degree matrices W' and W” respectively and substitute them into the segmentation model to calculate the subnet sets C' and C”;

[0052] Step 303: Merge subnet sets C' and C” using the overlap determination rule. If the overlapping area of ​​the closed regions constructed by each subnet element in the subnet set is greater than ΔC, then merge them to form a new subnet set C”'.

[0053] Step 304: Merge the overlapping subnet elements in region C”' to form the final road subnet set C”” = (G1, G2, ..., G m ), where m represents the number of road subnets in the road subnet set C”'.

[0054] Step 104: Classify the operating status of the road subnet according to the status indicators, define the road subnet in a congested state as a controlled subnet, and calculate the carrying capacity of each road subnet under congestion; the specific process includes:

[0055] For intersections and road segments in the road network, select appropriate status index I and define its value range [0, 100].

[0056] A polynomial fitting method is used to generate the traffic state-on-the-road vehicle relationship function curve.

[0057] Traffic state-on-the-way vehicle relationship function curve The traffic conditions are classified into three states: smooth, slow, and congested.

[0058] Define the road subnet that is in a congested state as a controlled subnet;

[0059] Calculate the state boundary values ​​of the road subnet to obtain the carrying capacity v of the controlled subnet. c .

[0060] In this embodiment of the disclosure, a polynomial fitting method is used to form the traffic state-on-the-road vehicle relationship function curve. Reference Figure 5 The specific calculation steps are as follows:

[0061] Step 401: Select and analyze traffic flow data for one week of the road network, and use Δt l , Δt l Calculate the time-series status index I for each day and each road subnetwork at intervals of ≥5 minutes, and obtain the number of vehicles on the road within the subnetwork v to form a dataset A. d =((v0,I0),(v1,I1)…(v n ,I n ), d=1,...,7, n=288 represents that there are 288 data points in a day, (v0,I0) represents that the status index corresponding to vehicle v0 is I0, where Δt l ≥5 minutes;

[0062] Step 402: For data set A d Sort the data by the number of vehicles en route from smallest to largest; if the values ​​of v are the same, sort them by time sequence to form a new data set A'. d ;

[0063] Step 403: For data set A' d The abnormal data is processed by identifying and replacing outliers; elements with the same value v are merged, and the I values ​​are averaged to form a new data set A. d ;

[0064] Step 404: For data set A” d A multinomial fitting method was used to fit the traffic state-on-the-road vehicle relationship function. in: I represents the status indicator, and x represents the number of vehicles on the road.

[0065] In this embodiment of the disclosure, the traffic state-on-the-way vehicle relationship function curve is characterized by feature classification, and the state is divided into three levels: smooth flow, slow flow, and congestion. Figure 6 As shown, the specific calculation steps are as follows;

[0066] Step 501: Based on one week's historical traffic data, obtain the minimum value v of the number of vehicles on the road. min and maximum value v max Construct the range of values ​​for the number of vehicles on the road, [v min 1.2*v max ];

[0067] Step 502: Construct a sample set (v) with an interval of Δv vehicles. min +Δv,v min +2*Δv, …,1.2*v max Substitute the values ​​into the relational function to calculate the traffic state index values, forming a data sample C = (v i ,I i ), i = 1, 2, 3…m;

[0068] Step 503: Using a sliding window Δv l Δv l ≥5, move the calculation data sample C to form a curvature data sample K = (K1, K2, ... K n );

[0069] Step 504: Analyze the curvature data samples using temporal clustering. Sort the curvature data from smallest to largest to form three state intervals: smooth flow, slow flow, and congestion, respectively [K1, K...]. s1 ), [K s1 ,K s2 ],(K s2 ,K n ].

[0070] In this embodiment of the disclosure, the bearing capacity range [v1, v2] is calculated. s1 ), [v s1,v s2 ],(v s2 ,v n This allows us to obtain the number of vehicles that can be carried by the road subnet under different congestion conditions.

[0071] Step 105: Based on the segmentation results and combined with the spatial characteristics of the road network, identify key nodes inside and outside the controlled subnet and implement control strategies.

[0072] In this embodiment of the disclosure, a set of controlled subnets in the road network is identified based on historical data. External entry nodes associated with the controlled subnets are calculated and sorted. Based on their correlation, the external entry nodes are stratified into restricted areas, buffer zones, and diversion zones (wherein, nodes with a correlation higher than a first preset threshold are classified as restricted areas, nodes with a correlation lower than a second preset threshold are classified as diversion zones, and others are classified as buffer zones; the first preset threshold is greater than the second preset threshold). The correlation of nodes such as intersections and road segments within the subnets is calculated to form a set of key nodes and non-key nodes (wherein, nodes with a correlation higher than a third preset threshold form a set of key nodes).

[0073] In this embodiment of the disclosure, in actual operation, vehicle trajectory data in the road network is acquired in real time, and subnet G is calculated. i On-the-go vehicle data v i Determine Δv = v i -v c When Δv approaches the threshold, the green light time in the restricted area is reduced, and the speed of vehicles in the buffer zone is decreased; in the diversion area, guidance information is issued, alternative routes are provided, and vehicles are guided to change their travel routes.

[0074] In this embodiment of the disclosure, the traffic status indicators of key nodes within the subnet are calculated in real time. With the goal of balancing the spatial carrying capacity of each node, intelligent diversion methods are used to quickly divert the flow associated with key nodes, and control the status indicators of each node within the optimized target range.

[0075] In one embodiment, a road network equilibrium control system based on big data is provided, including a data receiving module, a data processing module, a data service module, and a control optimization module.

[0076] The data receiving module is used to receive real-time traffic flow data collected by the road network front-end detection equipment.

[0077] The data processing module is used to preprocess real-time traffic data to ensure data quality and transmit the data to the data service module; it requests historical data, combines it with spatial road network data, analyzes and processes the data, calculates status indicators, and identifies traffic status.

[0078] Among them, the data service module is the system's data bus, responsible for data entry and exit; and realizes data interaction with other modules.

[0079] The control optimization module obtains real-time indicator data and control subnet data from the data module in real time, generates optimization control commands in real time, and sends them to the front-end control execution equipment.

[0080] Specific limitations regarding the big data-based road network balancing control system can be found in the limitations of the big data-based road network balancing control method described above, and will not be repeated here. Each module in the aforementioned big data-based road network balancing control system can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0081] In one embodiment, an electronic device is provided, including a memory for storing computer instructions; and a processor connected to the memory for executing the computer instructions in the memory, and implementing the above-described big data-based road network balancing control method when executing the computer instructions.

[0082] The equilibrium control method and system disclosed herein are beneficial for balancing the distribution of vehicles in the road network space and improving the traffic efficiency of the road network.

[0083] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention without departing from its spirit and scope should be included within the protection scope of the present invention.

Claims

1. A road network equilibrium control method based on big data, characterized in that, The method includes the following steps: Obtain and parse the spatial topology of the road network, obtain the attribute parameters of intersections and road segments, and obtain the layout of detection equipment on the road network, as well as the equipment type and detectable parameters; Based on the spatial topology of the road network, calculate the status indicators of each road network node within the road network; Based on the correlation between nodes in the road network, a clustering method is used to spatially segment the road network into different road subnetworks. The operating status of the road subnet is classified according to the status indicators, the controlled subnet is identified, and the carrying capacity of the controlled subnet is calculated. Based on the above segmentation results and combined with the spatial characteristics of the road network, key nodes inside and outside the controlled subnetwork are identified and control strategies are implemented. The method of using clustering to spatially segment the road network into different road subnetworks includes the following specific steps: Based on the crowded node set N S =(P S L S Construct a road network graph G = (P, L), where P represents all congested nodes (P1, P2, ..., Pn) in the road network. n For any two nodes in P, they are connected by an edge. The road network is bidirectional. i,j ≠w j,i The correlation matrices W' and W” are formed: Constructing a road subnet segmentation model: In the formula, the road network graph G is divided into k unconnected road subnets, and the set of road subnets is G = (G1, G2, ..., G...). k ), G i G j For the i-th and j-th road subnets, satisfying And G1∪G2∪…∪G k =G; W is W' or W”. For complement; Substituting the correlation matrices W' and W” into the road subnet segmentation model respectively, we obtain the subnet sets C' and C”; The subnet sets C' and C” are merged according to the overlap determination rule: for the closed regions constructed by each subnet element in the subnet set, if the overlapping area of ​​the two closed regions is greater than the set threshold ΔC, they are merged to form a new subnet set C”'. Merge the subnet elements with overlapping closed regions in the new subnet set C”' to form the final road subnet set C”” = (G1, G2, ..., G m ), where m represents the number of road subnets in the road subnet set C”'.

2. The road network equilibrium control method based on big data according to claim 1, characterized in that, The attribute parameters include: intersection shape, number of entrances and exits, lane type and number.

3. The road network equilibrium control method based on big data according to claim 1, characterized in that, The road network nodes include intersections and road segments. The status indicators of intersections include single or combined indicators such as saturation, average vehicle delay, and queue length. The status indicators of road segments include single or combined indicators such as travel time, travel speed, number of stops, and saturation.

4. The road network equilibrium control method based on big data according to claim 1, characterized in that, The method for calculating the correlation degree of each node in the road network is as follows: A two-layer network model is constructed to represent the road network, where the upper layer network represents the traffic flow assignment model and the lower layer network represents the spatial structure. Points are used to represent intersections and lines between points represent road segments. Among them, the lower-level network topology relationship N A = (R, L, α), where R is the set of intersection points; L is the set of road segments connecting upstream and downstream intersections; α is the static correlation degree, representing the spatial correlation between intersections within the road network; Among them, the upper-layer network topology relationship N B = (O, D, β), where O is the starting point; D is the destination point; and β is the dynamic correlation degree, which is influenced by factors including traffic flow and travel time. According to the different control periods throughout the day, within each control period, Δt l Slicing is performed at intervals, and the time slice with the most congested nodes in each control period is searched and used as the analysis period; where Δt l ≥15 minutes; Obtain the set of congested nodes N during the analysis period S =(P S L S , w), where w is the correlation degree between crowded nodes, P S For the congested intersection point set, L S The edge set of road segments between intersections; Based on the two-layer network model, the correlation calculation model is obtained: w i,j ＝α i,j *β i,j Among them, w i,j Let α be the correlation degree between road network nodes i and j. i,j For static correlation degree, a i,j =1 / d i,j d i,j β is the average connectivity distance between network nodes i and j; i,j For dynamic correlation, β i,j =Q ij / Q i +Q j Q ij Q represents the bidirectional traffic volume between two road network nodes i and j. i Q j Let i and j be the nominal traffic volumes for nodes i and j, respectively. The formula for calculating nominal traffic volume is as follows: In the formula, n is the number of critical nodes, and g i q represents the weight of the selected key node. i This represents the traffic flow corresponding to this key node.

5. The road network equilibrium control method based on big data according to claim 4, characterized in that, The process of classifying the operational status of road subnetworks based on status indicators, identifying controlled subnetworks, and calculating the carrying capacity of controlled subnetworks includes: For the intersection and road segment selection status index I in the road network, the value range is defined as [0, 100]; A polynomial fitting method is used to generate the traffic state-on-the-road vehicle relationship function curve. Traffic state-on-the-way vehicle relationship function curve The traffic conditions are classified into three states: smooth, slow, and congested. Define the road subnet that is in a state of congestion as a controlled subnet; Calculate the state boundary values ​​of the road subnet to obtain the carrying capacity v of the controlled subnet. c .

6. The road network equilibrium control method based on big data according to claim 5, characterized in that, The method employs a polynomial fitting approach to generate a traffic state-on-the-road vehicle relationship function curve. The specific process includes: We will analyze traffic flow data from one week of the road network, and use Δt as the metric. l To calculate the status index I for each day and each road subnetwork at intervals, and to obtain the number of vehicles on the road within each road subnetwork v, a dataset A is formed. d =((v0,I0),(v1,I1)…(v n ,I n ), d=1,...,7, n=288 represents that there are 288 data points in a day, (v0,I0) represents that the status index corresponding to vehicle v0 is I0, where Δt l ≥5 minutes; For data set A d Sort the data by the number of vehicles en route from smallest to largest; if the values ​​of v are the same, sort them by time sequence to form a new data set A'. d ; For data set A' d The abnormal data is processed by identifying and replacing outliers; elements with the same value v are merged, and the values ​​I are averaged to form a new data set A″. d ; For data set A″ d A multinomial fitting method was used to fit the traffic state-on-the-road vehicle relationship function. in: I represents the status indicator, and x represents the number of vehicles on the road.

7. The road network equilibrium control method based on big data according to claim 6, characterized in that, The relationship curve between traffic state and on-the-road vehicles The traffic conditions are classified into three ranges: smooth flow, slow flow, and congestion. The specific process includes: Based on a week's worth of historical traffic data, obtain the minimum value v of the number of vehicles on the road. min and maximum value v max The range of values ​​for the number of vehicles on the road is: [v min 1.2*v max ]; Construct a sample set (v) with an interval of Δv vehicles. min +Δv,v min +2*Δv, ..., 1.2*v max Substitute the traffic state-on-the-way vehicle relationship function into the data sample C = (v i ,I i ), i = 1, 2, 3, ..., m; With sliding window Δv l Δv l ≥5, move the calculation data sample C to form a curvature data sample K = (K1, K2, ... K n ); Temporal clustering was used to analyze the curvature data samples. The data were sorted from smallest to largest curvature to form three state intervals: smooth flow, slow flow, and congestion, which are [K1, K...]. s1 ), [K s1 ,K s2 ],(K s2 ,K n ]; This yields the bearing capacity intervals [v1, v2] corresponding to the three state intervals. s1 ), [v s1 ,v s2 ],(v s2 ,v n ].

8. The road network equilibrium control method based on big data according to claim 7, characterized in that, Based on the above segmentation results and combined with the spatial characteristics of the road network, key nodes inside and outside the controlled subnetwork are identified, and control strategies are implemented. The specific process includes: Identify the set of controlled subnets based on historical data, calculate the external entry nodes associated with the controlled subnets, and sort and classify them. Based on the correlation, external entry nodes associated with the controlled subnet are processed in layers to form restricted areas, buffer areas, and diversion areas; among them, nodes with a correlation higher than a first preset threshold are classified as restricted areas, nodes with a correlation lower than a second preset threshold are classified as diversion areas, and others are classified as buffer areas; the first preset threshold is greater than the second preset threshold. Real-time acquisition of vehicle location and trajectory data in the road network, and calculation of subnet G i On-the-go vehicle data v i ; Determine Δv = v i -v c When Δv approaches a preset threshold, the green light time in the restricted area is reduced, and the speed of vehicles in the buffer zone is decreased; in the diversion area, guidance information is issued, alternative routes are provided, and vehicles are guided to change their travel routes. Based on the correlation, the intersection and road segment nodes within the subnet are classified and processed to form a set of key nodes and a set of non-key nodes; among them, nodes with a correlation higher than the third preset threshold form the set of key nodes. Real-time calculation of traffic status indicators at key nodes, with the goal of balancing the spatial carrying capacity of each node, and intelligent diversion methods to guide the flow of traffic associated with key nodes, thereby controlling the status indicators of each node within the optimized target range.

9. A road network equilibrium control system based on big data, based on the method of any one of claims 1 to 8, characterized in that, The system includes a data receiving module, a data processing module, a data service module, and a control optimization module; The data receiving module is used to receive real-time traffic flow data collected by the road network front-end detection equipment; The data processing module is used to preprocess real-time traffic data and transmit the data to the data service module; at the same time, it requests historical traffic data, combines it with spatial road network data, analyzes and processes the historical traffic data, calculates status indicators, and identifies traffic status. The data service module, the data bus of the multidimensional system, is responsible for data inbound and outbound tasks. And enable data interaction tasks with other modules; The control optimization module is used to obtain real-time indicator data and control subnet data from the data processing module in real time, generate optimized control commands in real time, and send them to the front-end control execution device.

Images